apply_workbook_changeset

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the destructiveHint=true annotation, the description discloses the safe two-phase protocol (preview before commit), the guardrail role of assertions, and that snapshot creation defaults to enabled — all meaningful safety context. It does not contradict the annotation. It stops short of describing failure semantics, such as what happens when an assertion fails or the expected SHA mismatches.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is front-loaded with a crisp purpose sentence, then organized into operation format, assertion list, and workflow steps. The two enumeration lists are long but essential given zero schema coverage. Every sentence earns its place, though the overall density makes it slightly heavy even for a complex tool.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a destructive, 9-parameter tool with no output schema, the description covers the make-or-break knowledge: the two-phase protocol, what preview returns (token and SHA), and the full operation/assertion vocabularies. Gaps remain around sample_limit, snapshot_filepath, and the precise contents of the preview output beyond the token, so an agent still faces some uncertainty before the first call.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 0% schema description coverage, the description carries the full burden and largely delivers: it defines the operations payload shape ({"tool": "name", "args": {...}}), enumerates all 11 supported operation tool names, enumerates the 9 assertion types, and explains that changeset_token and expected_workbook_sha256 come from the preview response. It leaves sample_limit and snapshot_filepath unexplained, which prevents a 5.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The first sentence names the specific actions (preview or commit), the resource (a group of workbook mutations), and the distinguishing features (assertion-backed, single transaction). This clearly separates it from single-mutation siblings like write_data_to_excel or format_range, and from apply_workbook_repairs, which addresses a different concept entirely.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives explicit workflow direction: 'Call preview first. Commit the identical plan with expected_workbook_sha256 and changeset_token from that preview.' This is strong when-to-use guidance for the two modes. However, it does not explicitly state when to prefer this batching tool over calling individual mutation tools directly, nor does it name an alternative.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.